How to Work with Content Stream Operators in Java

Access the Content Stream

Each Page in a Document holds its graphical content in a content stream. The stream contains a sequence of PDF operators that define text positioning, drawing commands, and resource references. Read the raw content stream bytes from a page:

try (Document doc = new Document("input.pdf")) {
    PageCollection pages = doc.getPages();
    Page page = pages.get(1);
    byte[] contentBytes = page.getContents().toByteArray();
    System.out.println("Content stream size: " + contentBytes.length + " bytes");
}

Parse with ContentStreamParser

ContentStreamParser tokenizes and parses a content stream into operator objects, making it possible to inspect each drawing command in the stream:

try (Document doc = new Document("input.pdf")) {
    PageCollection pages = doc.getPages();
    Page page = pages.get(1);
    byte[] contentBytes = page.getContents().toByteArray();
    ContentStreamParser parser = new ContentStreamParser(contentBytes);
    // Iterate over parsed operators
}

Build with ContentStreamBuilder

ContentStreamBuilder creates new content streams for embedding in a Page. Use ContentStreamBuilder when constructing page content programmatically rather than modifying an existing stream. After building the stream, assign it back to the page and save the document.

Operator Reference

PDF content stream operators are defined in ISO 32000-1:2008, Table A.1. Common operators include:

• Text operators: BT, ET, Tf, Tj
• Graphics state operators: q, Q, cm
• Path operators: m, l, h, f, S